conv.

All stories
Active · 38h

Humanising LLM Outputs Is Dumb

Humanising LLM Outputs Is Dumb
kuber.studio

Conversation activity · last 39 hours peak 3/hr

Peak 3 items in one hour at Aug 10, 8 AM; 16 items over 39 hours Aug 10, 8 AM — 3 items · Press 2, Hacker News 1Aug 10, 9 AM — no itemsAug 10, 10 AM — no itemsAug 10, 11 AM — no itemsAug 10, 12 PM — 2 items · Hacker News 2Aug 10, 1 PM — 2 items · Hacker News 2Aug 10, 2 PM — 2 items · Hacker News 2Aug 10, 3 PM — 1 item · Hacker News 1Aug 10, 4 PM — 1 item · Hacker News 1Aug 10, 5 PM — 1 item · Mastodon 1Aug 10, 6 PM — no itemsAug 10, 7 PM — no itemsAug 10, 8 PM — no itemsAug 10, 9 PM — 1 item · Hacker News 1Aug 10, 10 PM — no itemsAug 10, 11 PM — 2 items · Hacker News 2Aug 11, 12 AM — no itemsAug 11, 1 AM — no itemsAug 11, 2 AM — no itemsAug 11, 3 AM — no itemsAug 11, 4 AM — 1 item · Hacker News 1Aug 11, 5 AM — no itemsAug 11, 6 AM — no itemsAug 11, 7 AM — no itemsAug 11, 8 AM — no itemsAug 11, 9 AM — no itemsAug 11, 10 AM — no itemsAug 11, 11 AM — no itemsAug 11, 12 PM — no itemsAug 11, 1 PM — no itemsAug 11, 2 PM — no itemsAug 11, 3 PM — no itemsAug 11, 4 PM — no itemsAug 11, 5 PM — no itemsAug 11, 6 PM — no itemsAug 11, 7 PM — no itemsAug 11, 8 PM — no itemsAug 11, 9 PM — no itemsAug 11, 10 PM — no items 3 items · 8 AM
4 PMAug 118 AM4 PMnow · 11:52 PM

Clustered from 16 items across 3 sources. Not yet parsed — the coverage below is the raw record.

Press coverage 1

Social posts 2

Voices from the web unedited

  • > The problem is that these instructions are not applied after the model has finished doing the work, it becomes part of the same work - If you tell an agent to use short sentences, avoid jargon, never overwhelm you and only include the most important details, you are asking it to continuously compress its output into a lower-bandwidth format.>…

    wren6991Hacker News1d agoview on Hacker News ↗
  • Humanising LLM Outputs Is Dumb https:// news.ycombinator.com/item?id=4 9243474 # hackernews # tech

    HackerNewsBot@m.einverne.infoMastodon · fosstodon.org1d agoview on Mastodon ↗
  • I agree.I've come to believe this is also a side effect of the desire for less (/goal: no) human in the loop on the part of the people driving all this capex spend. I think if you actually want to manually review output there will be a moment where you will actually want a separate interface to a stupider or "simpler" model. I suspect sometimes…

    godwinson__4-8Hacker News1d agoview on Hacker News ↗
  • You ever read a work of literature with such flowery language that right after you've read a paragraph, you pause and realize you have no clue what you actually read, only to read the paragraph maybe a second or third time and have your mind space out again and again on each successive attempt?Yeah, for me, that's what parsing huge volumes of…

    XcelerateHacker News1d agoview on Hacker News ↗
  • People talk to their pets, plants. Their cars or other inanimate objects even. Anthropomorphizing stuff around us is a natural thing to do. It might be irrational but it just fits the way our brains work. The natural way to interact with an LLM is to pretend it's just another person. And LLMs are of course very good at emulating that to the point…

    jillesvangurpHacker News18h agoview on Hacker News ↗
  • And on the "input" side, one thing that used to improve google search result was to write like you are talking to a robot. "Ruby on rails http header set function". As opposed to "how do I set header in ruby?" Then you have to page through results until you find something specific to rails.Now, the second example is the only thing that works…

    firefoxdHacker News1d agoview on Hacker News ↗
  • I don't like it when the LLM tries to be my friend. My general prompt (a work in progress) is this. I wonder what other people use."Answer impersonally, objectively and analytically, without undue friendliness or enthusiasm. Use an engineering style response: concise, factual, and complete. Do not speak in the first person. Do not promote…

    7402Hacker News1d agoview on Hacker News ↗
  • The effect you describe reminds me of reading Edward W. Said's "Orientalism" when I was younger. Fable suddenly started with this kind of lingo, iirc, and Opus 5 sounds exactly the same. Tin foil: it's ultimately a vendor lock-in strategy, you'll get the best results with agents from the same tribe, others will trip over the mountain of…

    whythismattersHacker News1d agoview on Hacker News ↗
  • Well, what do you expect? LLMs are trained on blithering, mostly from web sites. So you get blithering out.There's an important point in the article, that forcing a style onto an LLM is lossy. Although he doesn't seem to mention it, forcing a style may result in the insertion of new blithering, possibly made up as a hallucination.

    AnimatsHacker News1d agoview on Hacker News ↗
  • It's usually a pretty clear case of reaching for statements that both sound impressive while also being broad and vague enough they are less likely to be factually wrong. In this regime, being difficult to parse is actually part of the performance, because it prevents the user from being able to spot a clear error.

    ip26Hacker News23h agoview on Hacker News ↗